FreeBSD Continues Push Toward Deprecating 32-bit Platforms

[phoronix]

Phoronix: FreeBSD Continues Push Toward Deprecating 32-bit Platforms

As noted in prior Phoronix articles for months, FreeBSD 14 is likely to be the last for supporting prominent 32-bit systems. On the FreeBSD mailing list more details on the FreeBSD 32-bit deprecation process was posted this week...

https://www.phoronix.com/news/FreeBSD-32-bit-Deprecation-Plan

[varikonniemi]

Every distro should make one last 32bit LTS release and then just drop it. In 6 years or so when it becomes EOL there is certainly no need to still use such processors in production. And hobbyist projects can then support them as long as the kernel supports it.

[jaypatelani]

Every distro should make one last 32bit LTS release and then just drop it. In 6 years or so when it becomes EOL there is certainly no need to still use such processors in production. And hobbyist projects can then support them as long as the kernel supports it.

Don't worry NetBSD has their back : https://wiki.netbsd.org/ports/i386/

[ayumu]

Corporate priorities. For a while now, Freebsd only exists to serve corpos.

[jaypatelani]

Corporate priorities. For a while now, Freebsd only exists to serve corpos.

Ironically Linux has been backbone to serve most corpos. RedHat to Google.

[ayumu]

Ironically Linux has been backbone to serve most corpos. RedHat to Google.

Yup. And just recently they made a lot of our older (but still in use) hardware much less useful, by removing PCMCIA support.

[Setif]

Every distro should make one last 32bit LTS release and then just drop it. In 6 years or so when it becomes EOL there is certainly no need to still use such processors in production.

Will you pay them from your pocket?, Or Is it part of an agreement /contract we don't know about?
No one could dictate on someone who's offering something for free when could he stop offering it.

[User29]

Every distro should make one last 32bit LTS release

Playing with other people's money and time.... Should could would....

[caligula]

Yup. And just recently they made a lot of our older (but still in use) hardware much less useful, by removing PCMCIA support.

The hardware is now 15-35 years old. It will still take years before you start considering upgrading to distros that don't have any 32-bit support. It will still be possible to upgrade the userspace apps separately. The new kernels typically don't provide any new features for 15+ year old hardware.

[endrebjorsvik]

Yup. And just recently they made a lot of our older (but still in use) hardware much less useful, by removing PCMCIA support.

I completely agree with kernel maintainers that PCMCIA hardware is obsolete today. PCMCIA was succeeded by ExpressCard in 2003. Around that time we also saw a significant improvement in CPU efficiency. The horribly inefficient Pentium 4 was replaced by the vastly improved Pentium M/Dothan/Core/Yonah family of CPUs. Dothan was released in 2003 and Yonah in 2006. I consider anything prior to those CPU families completely obsolete. They are burning lots of power with very little performance. The Pentium M was not a huge commercial success, so there are mainly Core CPUs left from that era. Given that those were launched in 2006, I firmly believe that most of them shipped with ExpressCard instead of PCMCIA.

There might be a few odd ones out there with Core and PCMCIA, but I do not believe that it is worth spending precious kernel development time on those. You still have the possibility of running older kernels on that hardware. Especially LTS stuff like RHEL 8, which is supported until 2029.